Sit back and get comfortable as I try to explain whats going on with my car.

First let me take you back a couple of weeks ago to when I was at Island dragway doing some shakedown runs. I was having a hard time coming out of the whole, ususally I pull 1.7s - 60's with a quick slip off the 2-step but my best that day was a mid 1.8 with a looong slip of the clutch. I wasn't able to figure out what was happening until I looked at my boost log and noticed that I was only making 1-2psi instead of the 5-6psi I usually see. Also at the top of first gear I was making much less boost then normal, usually first gear has the highest boost spike out of the entire run. I was only making around 18psi instead of 22-23 like normal.

Below is screen shot of the log showing what I'm talking about. the top one is how I usually launch, notice boost is good while launching and spikes high at the end of the gear. The bottom one is the one I'm talking about with less boost all over.





I thought that maybe my MBC wasn't working correctly so I switched it out for another good unit that I know works but that didn't fix the problem. Then I tried switching how the MBC was hooked up, its a Tial 38mm WG, but that didn't have make a difference either.

THis doesn't just happen when I try to launch on the strip, but also during normal driving on the street. Say my boost is set to 19psi, if I go WOT the boost might only climb to 15-16psi and just sit there. Then all of a sudden it will SPIKE up to 19psi where it should be. Below is another log showing what I'm talking about.





Notice I'm WOT and boost starts to settle around 14-15psi and then all of a sudden it jumps to 19 where it should be. It doesn't do it consistantly, I'd say maybe 75-80% of the time I go WOT. I don't believe its the boost contoller at all, I'm leaning towards the WG being bad but I want to see what other people think.


____________________________________________



Ok so that was just the one problem. Its not why I made this post but I think it might be related to the other more serious issue I'm having now. This new problem just started to manifest itself last night as I was out trying to diagnose my boost issue described above.

What just started happening is now when my boost spikes, like I described above, my A/F ratio is spiking now too instead of dropping like it normally would do. The A/F only spikes though when the boost spikes and so far I can't tune around it. And as an awesome side effect when the A/F spikes its also giving my SERIOUS knock counts, yah!. I have one log w/32 counts of knock and another with 20 counts. Below is another log showing the boost spike and the A/F spike from last night. You'll see they follow each other perfectly.



Usually when load increases I have my maps scaled to richen everything up and add timing. This has worked great until this boost thing started happening, now its like the ECU has a mind of its own. I'm hoping someone here has seen this before and maybe has an idea of what is going on.

A couple of possible things I think might be the problem are:
faulty WG - although ususally when they fail, which isn't very often, they stay closed instead of open.
bad gas - but I've been filling up at the same 2 stations for months and never had an unusual knock issue. besides that doesn't explain the boost problem.
bad walbro - its pretty new, maybe 20k old, but that also doesn't explain the boost issure.

For reference my setup is a 100% stock block 04 GSR w/ a gt35r, 1000cc injectors, tail 38mm WG, walbro fuel pump, 3" exhaust & obviously ECUFlash. I have zero boost leaks and I'm only running 22psi on pumpgas.

 

Did you mess with any of the lines coming off of the FPR by any chance? Did you check those for any tears or leaks? Is that where your boost gauge is T'd into?

Also, did you mess with the MAF in any way? Is that still holding up well?

That's definitely an interesting condition you have there. It may actually be two separate problems, though. The AFR rising with boost seems to obviously be a fuel delivery problem or airflow reading problem. That's why I asked about the FPR lines....it almost looks like the FPR isn't raising the FP with the increased boost. It also looks like it could be a fuel pump going bad.

But, neither of those really correlates with your logged boost. If your boost sensor is T'd into the FPR line, it may be somewhat related, but not the way it is shown in your graphs above.

I'll keep thinking.


EDIT: Also, for your Tial external gate, since your are using a MBC and not an EBC, I am assuming that you have the top port open and the MBC hooked to the side port?


Eric

 

Good observations Eric. last night when this nasty lean condition started happening I did find the line to the FPR had popped off. I was really hoping that was going to be the fix... I didn't drive it around too much after I zip-tied the line back on cause it was getting late but I did get on it a few times, one of those times the A/F spiked.

to answer your other questions; I don't have any other lines 'T' into the FPR signal line. my boost signal for the map sensor/boost gauge comes off a port on the back of the mainfold.

the MAF honeycomb seems to be holding up well, I am only running 23psi max but I guess its not enough CFM to cause a problem, yet.

On the Tial WG I was using just the bottom port, thats how I would always run it. Once this started happening though I switched it to the two port method to see if that would fix it. I also tried a different MBC using both methods, no difference.

Looking at my EVOScan logs from last night I noticed that the IPW is getting very high, between 16-18ms, when the A/F spikes. duty cylces are aound 90-95% when this happens. what would cause the injectors to do this?

 

So you did a full leak test with something that pressurised the system or just check the clamps? It seems an awful lot like there has to be a boost leak somewhere or your WG took a crap on you. I will figure that since you check for leaks, its probably a WG problem. I have a 35mm from back in the day that split a diaphragm all on its own one day for no apparent reason. Take it off and take it apart after you have exhausted (no pun intended) all other avenues.

 

boost leak test was done with air, not the 'eyeball' method.

So I realized that the last time I looked at the MAF was a couple of days ago, before all this lean stuff starting happening. So I went out to the garage and guess what I found...




edit... sorry for the huge pic.

 

Well try to straighten that out, See if it fixes it, i dont know if that part open could cause a lower pressure system then the already low pressure in the inlet to the MAF. And it might be throwing off the reading?

 

I wonder can be used to reinforce the stock maf honeycomb without affecting its readings.

 

Welcome to the MAF sucking club.

What I noticed with the maf screen is as long as the mesh is good in front of the sensors there should be no hiccups in the power band. But you never know. Might as well straighten it out and see if that cures the problem. What type of filter are you running. When my screen got sucked in I had a K/N filter on there. Now that the car has a HKS type foam filter my current MAF has lasted longer than my previous one. But after 3k or so on a brand new MAF, you can see the start of some distortion in the mesh.

 

the honeycomb part directly in front of the sensor was 'ok', I took it out and straightened it but I'm not going to reuse it. I'm too worried its going to eventually wind up hitting the compressor wheel.

I was using a K&N cone filter and running 20psi on a 35r when this happened.

I don't know if the honeycomb thing was the fix to my fueling problem either. I pulled the screen out of the MAF and drove the car for a little last night, my logs show that A/F is STILL spiking. It might have something to do with the screen not being in there at all, but the car felt fine everywhere else in the RPM band. I know people say removing the MAF is a big no-no but other then the condition I was experiencing above it seems to drive fine. So until I get a new MAF the jury is out on if the screen is the culprit or not.

Today I'm pulling the 35r and going to reinstall my stock setup until I can figure out what is going on with the boost and with the fuel. If I reinstall the stock turbo and boost is fine again then I know its the Tial WG and I can concentrate on getting the fuel issue worked out. But if the car still has the same problem its going to be back to square one...

anyone have more advice?

 

Have you tested the FPR yet? You have to measure fuel pressure with the vacume line on and off, reading hsould be higher with the line off (vacume port seeing atmospheric pressure rather than vacume).

Sounds like you are not getting enough fuel under load, so walbro and fpr would be something to look at, then again if the maf reading is way off........

 

I would say that the MAF was definitely part of the problem if not all of it.

With the honeycombs basically removed like they were in your MAF around the outside of the sensor, it was like running a hacked MAF in a DSM. That was an old trick to get more airflow out of the MAF. But, you would run a lot leaner under boost, so with DSMLink you could alter the MAF table to calibrate it back.

The reason it's not lean at lower RPMs is because the air is moving relativelty slow and the screen isn't much of a restriction. But under high flow/high boost, a lot of air is now being sucked in quickly and it will find the easiest way through the MAF, which is where the big holes were. Any extra air that took that path wasn't accounted for by the calibration of the MAF sensor, resulting in a leaner than expected A:F ratio.

EDIT: Actually, I looked up the hacked MAF settings in the DSMLink forums because I re-read what I wrote and I thought it sounded a little off. Well, my general reasoning was right...it will make you run a lot leaner, but you will run even leaner at lower Hz reading that at higher. I'm guessing that your fuel trims were just helping you out at idle and cruise, and when you went WOT and changed to open-loop, then your trims couldn't help anymore and you ran lean because of the 'hacked' MAF.

Were your LTFTs (Lo and Mid) very positive?

Just for the sake of information, these were the values that you would have to use for a hacked 2G MAF:

Hz adjustment
50 +48%
150 +30%
250 +25%
400 +22%
800 +17%
1200 +13%
1600 +13%
2000 +13%
2400 +13%

So, you can see, that on the 2G MAF, removing the honeycombs would make you run about 13% leaner at higher RPMs.


Eric

 

I agree that the MAF is probably the main reason for the leaning out. I have a few PMs out there so I should have another one soon.

That solves one problem but I'm still having that boost issue which I can't seem to figure out. This afternoon I swapped out my 35r setup for the stock turbo and the car is still boosting the same.

This is very strange, I've never seen anything like this before...

 

come on... none of the big dogs have an idea what this could be???

 

I don't think we can adequately answer your question until you get some logs with your MAF fixed..


Also your boost issue.. Check your DV/BOV.. Theres a distinct possibility that your leaking until the engine is ingesting enough air that the boost starts building up again and not forcing the valve open.. It sounds contrary to common sense, but if you can't find anything else wrong, and your using an MBC, then you either have an issue with the MBC itself, or the DV may not be sealing completely until the pressure and vacuum source to the DV is adequate..

What are you using for a pressure source to your actuator, and to your DV?

 

Jack, I think the lean spike issue is resolved, I'm just waiting on a new MAF to confirm.

I think your on to something w/ the DV/BOV comment. right before all this started happening I switched from my JDM/MR bov to a Forge unit (the one with the dsm flange) and buschur UICP. I was getting some leaking at the base of the BOV, where the valve body meets the flange, when I pressure tested the system so I JB welded it and about that time is when all this boost stuff started happening.

The pressure source for the BOV is the stock line which comes off the manifold, nothing else is tapped into it. The pressure source for the WG comes directly from the compressor cover on the turbo, it's a dedicated line also.